>>> On 13.07.18 at 11:04, <aisa...@bitdefender.com> wrote:
> Changes since V10:
> - Add memset to 0 for ctxt.
Why? What's wrong with ...
> --- a/xen/arch/x86/cpu/mcheck/vmce.c
> +++ b/xen/arch/x86/cpu/mcheck/vmce.c
> @@ -349,6 +349,20 @@ int vmce_wrmsr(uint32_t msr, uint64_t val)
> return ret;
> }
>
> +static int vmce_save_vcpu_ctxt_one(struct vcpu *v, hvm_domain_context_t *h)
> + {
> + struct hvm_vmce_vcpu ctxt;
> +
> + memset(&ctxt, 0, sizeof(ctxt));
struct hvm_vmce_vcpu ctxt = {};
instead? Or even pulling all of this ...
> + ctxt.caps = v->arch.vmce.mcg_cap;
> + ctxt.mci_ctl2_bank0 = v->arch.vmce.bank[0].mci_ctl2;
> + ctxt.mci_ctl2_bank1 = v->arch.vmce.bank[1].mci_ctl2;
> + ctxt.mcg_ext_ctl = v->arch.vmce.mcg_ext_ctl;
... into the initializer, which will make sure padding fields are zero.
I can see that a memset() is warranted in _some_ cases, but not
uniformly.
Jan
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel
